#!/bin/bash
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn
# gg_monitor.sh
# from:www.itpux.com.qq113257174.wx:itpux-com
# web: http://www.fgedu.net.cn

# 配置
GG_HOME=”/oracle/goldengate”
LOG_FILE=”$GG_HOME/monitor/gg_monitor_$(date +%Y%m%d).log”
ALERT_FILE=”$GG_HOME/monitor/gg_alert_$(date +%Y%m%d).log”

# 创建监控目录
mkdir -p $GG_HOME/monitor

# 记录日志
log_message() {
echo “$(date +”%Y-%m-%d %H:%M:%S”): $1″ >> $LOG_FILE
}

# 记录告警
alert_message() {
echo “$(date +”%Y-%m-%d %H:%M:%S”): $1″ >> $ALERT_FILE
# 发送告警邮件
echo “$1” | mail -s “GoldenGate Alert” admin@fgedu.net.cn
}

# 检查进程状态
check_processes() {
log_message “检查GoldenGate进程状态…”

PROCESS_STATUS=$($GG_HOME/ggsci << EOF info all EOF ) log_message "进程状态: $PROCESS_STATUS" # 检查是否有进程异常 if echo "$PROCESS_STATUS" | grep -E "ABENDED|STOPPED" > /dev/null; then
alert_message “发现异常进程状态: $PROCESS_STATUS”
fi
}

# 检查复制延迟
check_lag() {
log_message “检查复制延迟…”

# 检查Extract进程延迟
EXTRACT_LAG=$($GG_HOME/ggsci << EOF lag extract ext01 EOF ) log_message "Extract延迟: $EXTRACT_LAG" # 检查Replicat进程延迟 REPLICAT_LAG=$($GG_HOME/ggsci << EOF lag replicat rep01 EOF ) log_message "Replicat延迟: $REPLICAT_LAG" # 检查延迟是否超过阈值 if echo "$EXTRACT_LAG" | grep -E "[0-9]{2}:[0-9]{2}:[0-9]{2}" > /dev/null; then
EXTRACT_LAG_HOURS=$(echo “$EXTRACT_LAG” | grep -E “[0-9]{2}:[0-9]{2}:[0-9]{2}” | awk ‘{print $3}’ | cut -d: -f1)
if [ “$EXTRACT_LAG_HOURS” -gt 1 ]; then
alert_message “Extract延迟超过1小时: $EXTRACT_LAG”
fi
fi

if echo “$REPLICAT_LAG” | grep -E “[0-9]{2}:[0-9]{2}:[0-9]{2}” > /dev/null; then
REPLICAT_LAG_HOURS=$(echo “$REPLICAT_LAG” | grep -E “[0-9]{2}:[0-9]{2}:[0-9]{2}” | awk ‘{print $3}’ | cut -d: -f1)
if [ “$REPLICAT_LAG_HOURS” -gt 1 ]; then
alert_message “Replicat延迟超过1小时: $REPLICAT_LAG”
fi
fi
}

# 检查Trail文件
check_trail_files() {
log_message “检查Trail文件…”

TRAIL_DIR=”$GG_HOME/dirdat”
TRAIL_FILES=$(ls -la $TRAIL_DIR)
log_message “Trail文件: $TRAIL_FILES”

# 检查Trail文件大小
LARGE_TRAIL_FILES=$(find $TRAIL_DIR -name “*.r*” -size +1G)
if [ -n “$LARGE_TRAIL_FILES” ]; then
alert_message “发现大于1GB的Trail文件: $LARGE_TRAIL_FILES”
fi

# 检查Trail文件数量
TRAIL_COUNT=$(ls -la $TRAIL_DIR | wc -l)
if [ “$TRAIL_COUNT” -gt 100 ]; then
alert_message “Trail文件数量过多: $TRAIL_COUNT”
fi
}

# 检查系统资源
check_system_resources() {
log_message “检查系统资源…”

# 检查CPU使用情况
CPU_USAGE=$(top -b -n 1 | grep “Cpu(s)”)
log_message “CPU使用情况: $CPU_USAGE”

# 检查内存使用情况
MEM_USAGE=$(free -m)
log_message “内存使用情况: $MEM_USAGE”

# 检查存储空间
DISK_USAGE=$(df -h)
log_message “存储空间使用情况: $DISK_USAGE”

# 检查磁盘空间是否不足
LOW_DISK=$(df -h | grep -E “[0-9]{2}%” | awk ‘{if ($5+0 > 90) print $0}’)
if [ -n “$LOW_DISK” ]; then
alert_message “磁盘空间不足: $LOW_DISK”
fi
}

# 主函数
main() {
log_message “开始GoldenGate监控”
check_processes
check_lag
check_trail_files
check_system_resources
log_message “GoldenGate监控完成”
}

# 执行主函数
main

3.3 基于Enterprise Manager的监控